6biCSrSSKJs Jr SSKJr SSKJr SSK J r SSK J r SSK Jr SSK Jr SS KJr S r\S -r\S -r\S -r\S-r\S-r\S-r\ "5rSrSrSrS&Sjr\"SS5S&Sj5r\"SS5S&Sj5r\"SS5S&Sj5r \"S5S'Sj5r!\"S 5S(S!j5r"\RFRIS"\RJ\RLS#9\!l\RDR\"lS$r'\("\S%\R\'-5 \("\S%\R\'-5 \("\ S%\ R\'-5 g))zDenseNet models for TF-Keras. Reference: - [Densely Connected Convolutional Networks]( https://arxiv.org/abs/1608.06993) (CVPR 2017) N)backend)imagenet_utils)training)VersionAwareLayers) data_utils) layer_utils) keras_exportzFhttps://storage.googleapis.com/tensorflow/keras-applications/densenet/1densenet121_weights_tf_dim_ordering_tf_kernels.h57densenet121_weights_tf_dim_ordering_tf_kernels_notop.h51densenet169_weights_tf_dim_ordering_tf_kernels.h57densenet169_weights_tf_dim_ordering_tf_kernels_notop.h51densenet201_weights_tf_dim_ordering_tf_kernels.h57densenet201_weights_tf_dim_ordering_tf_kernels_notop.h5c d[U5H n[USUS-[US-5-S9nM" U$)zA dense block. Args: x: input tensor. blocks: integer, the number of building blocks. name: string, block label. Returns: Output tensor for the block. _blockname)range conv_blockstr)xblocksris \/home/james-whalen/.local/lib/python3.13/site-packages/tf_keras/src/applications/densenet.py dense_blockr?s66] q"4(?SQZ#? @ Hcv[R"5S:XaSOSn[RUSUS-S9"U5n[R SUS-S 9"U5n[R [ [R"U5UU-5SS US -S 9"U5n[RS S US-S9"U5nU$)zA transition block. Args: x: input tensor. reduction: float, compression rate at transition layers. name: string, block label. Returns: output tensor for the block. channels_lastr>_bnaxisepsilonrrelu_relurF_convuse_biasr_poolstridesr) rimage_data_formatlayersBatchNormalization ActivationConv2Dint int_shapeAveragePooling2D)r reductionrbn_axiss rtransition_blockr:Os,,./AaqG!! hTE\ "  A &tg~6q9A  G  a  )I 56  G^     A 14'>B1EA Hrc[R"5S:XaSOSn[RUSUS-S9"U5n[R SUS-S 9"U5n[R S U-SS US -S 9"U5n[RUSUS-S9"U5n[R SUS-S 9"U5n[R USSS US-S9"U5n[R X2S-S9"X/5nU$)zA building block for a dense block. Args: x: input tensor. growth_rate: float, growth rate at dense layers. name: string, block label. Returns: Output tensor for the block. r r!rr"_0_bnr$r'_0_relurF_1_convr*_1_bn_1_relusame_2_conv)paddingr+r_concat)r%r)rr0r1r2r3r4 Concatenate)r growth_raterr9x1s rrris-,,./AaqG  " " hTG^ #  B   6y(8  9" =B  KU 1A   B  " " hTG^ #  B   6y(8  9" =B QTI=M   B Y.>?HA Hrc US;d9[RRRU5(d [ S5eUS:XaU(aUS:wa [ S5e[ R "USS[R"5UUS9nUc[RUS 9nO1[R"U5(d[RX4S 9nOUn[R"5S :XaS OS n [RSS9"U5n [RSSSSSS9"U 5n [RU SSS9"U 5n [RSSS9"U 5n [RSS9"U 5n [R!S SSS9"U 5n [#XSS S9n [%U S!S"S9n [#XS S#S9n [%U S!S$S9n [#XSS%S9n [%U S!S&S9n [#XS S'S9n [RU SS(S9"U 5n [RSSS9"U 5n U(aJ[R'S)S9"U 5n [ R("Xr5 [R+XgS*S+9"U 5n O?US,:Xa[R'S)S9"U 5n OUS-:Xa[R-S.S9"U 5n Ub[.R0"U5n OUn U/S/Q:Xa[2R4"XS0S9n OQU/S1Q:Xa[2R4"XS2S9n O3U/S3Q:Xa[2R4"XS4S9n O[2R4"XS5S9n US:XaU(alU/S/Q:Xa[6R8"S6[:S7S8S99n OU/S1Q:Xa[6R8"S:[<S7S;S99n OU/S3Q:Xa8"S<[>S7S=S99n OkU/S/Q:Xa[6R8"S>[@S7S?S99n OGU/S1Q:Xa[6R8"S@[BS7SAS99n O#U/S3Q:Xa8"SB[DS7SCS99n U RGW 5 U $UbU RGU5 U $)Da Instantiates the DenseNet architecture. Reference: - [Densely Connected Convolutional Networks]( https://arxiv.org/abs/1608.06993) (CVPR 2017) This function returns a TF-Keras image classification model, optionally loaded with weights pre-trained on ImageNet. For image classification use cases, see [this page for detailed examples]( https://keras.io/api/applications/#usage-examples-for-image-classification-models). For transfer learning use cases, make sure to read the [guide to transfer learning & fine-tuning]( https://keras.io/guides/transfer_learning/). Note: each TF-Keras Application expects a specific kind of input preprocessing. For DenseNet, call `tf.keras.applications.densenet.preprocess_input` on your inputs before passing them to the model. `densenet.preprocess_input` will scale pixels between 0 and 1 and then will normalize each channel with respect to the ImageNet dataset statistics. Args: blocks: numbers of building blocks for the four dense layers. include_top: whether to include the fully-connected layer at the top of the network. weights: one of `None` (random initialization), 'imagenet' (pre-training on ImageNet), or the path to the weights file to be loaded. input_tensor: optional TF-Keras tensor (i.e. output of `layers.Input()`) to use as image input for the model. input_shape: optional shape tuple, only to be specified if `include_top` is False (otherwise the input shape has to be `(224, 224, 3)` (with `'channels_last'` data format) or `(3, 224, 224)` (with `'channels_first'` data format). It should have exactly 3 inputs channels, and width and height should be no smaller than 32. E.g. `(200, 200, 3)` would be one valid value. pooling: optional pooling mode for feature extraction when `include_top` is `False`. - `None` means that the output of the model will be the 4D tensor output of the last convolutional block. - `avg` means that global average pooling will be applied to the output of the last convolutional block, and thus the output of the model will be a 2D tensor. - `max` means that global max pooling will be applied. classes: optional number of classes to classify images into, only to be specified if `include_top` is True, and if no `weights` argument is specified. classifier_activation: A `str` or callable. The activation function to use on the "top" layer. Ignored unless `include_top=True`. Set `classifier_activation=None` to return the logits of the "top" layer. When loading pretrained weights, `classifier_activation` can only be `None` or `"softmax"`. Returns: A `keras.Model` instance. >NimagenetzThe `weights` argument should be either `None` (random initialization), `imagenet` (pre-training on ImageNet), or the path to the weights file to be loaded.rJzWIf using `weights` as `"imagenet"` with `include_top` as true, `classes` should be 1000r) default_sizemin_size data_formatrequire_flattenweights)shape)tensorrRr r!r)r!r!rT)rD@r,Fz conv1/conv)r/r+rr"zconv1/bnr$r'z conv1/relur)rrrWpool1r.rconv2g?pool2conv3pool3conv4pool4conv5bnavg_pool predictions) activationravgmaxmax_pool  densenet121rhrirr densenet169rhri0r densenet201densenetr models 9d60b8095a5708f2dcce2bca79d332c7) cache_subdir file_hashr d699b8f76981ab1b30698df4c175e90br 1ceb130c1ea1b78c3bf6114dbdfd8807r 30ee3e1110167f948a6b9946edeeb738r b8c4d4c20dd625c148057b9ff1c1176br c13680b51ded0fb44dff2d8f86ac8bb1)$tfiogfileexists ValueErrorrobtain_input_shaperr0r1Inputis_keras_tensor ZeroPadding2Dr4r2r3 MaxPooling2Drr:GlobalAveragePooling2Dvalidate_activationDenseGlobalMaxPooling2Drget_source_inputsrModelrget_fileDENSENET121_WEIGHT_PATHDENSENET169_WEIGHT_PATHDENSENET201_WEIGHT_PATHDENSENET121_WEIGHT_PATH_NO_TOPDENSENET169_WEIGHT_PATH_NO_TOPDENSENET201_WEIGHT_PATH_NO_TOP load_weights)r include_toprQ input_tensor input_shapepoolingclassesclassifier_activation img_inputr9rinputsmodel weights_paths rDenseNetrsNT ) )RUU[[-?-?-H-H <  *D 1  !33--/# KLL{L3 &&|44 L LI$I,,./AaqG%56yAA b!Q\ J1MA!! hZ "  A &|4Q7A%56q9AAqw7:AAayw/ACg.AAayw/ACg.AAayw/ACg.AAayw/A!!wt!LQOA&v.q1A  ) )z ) :1 =**+@J LL M    e --:->qAA  ))z):1=A..|< v}= ? "v}= ? "v}=vz:* ()22G+!)@   ?*)22G+!)@   ?*)22G+!)@   ()22M2!)@   ?*)22M2!)@   ?*)22M2!)@   <( L   7# Lrz'keras.applications.densenet.DenseNet121zkeras.applications.DenseNet121c *[/SQUUUUUUU5$)z*Instantiates the Densenet121 architecture.rgrrrQrrrrrs r DenseNet121rZ)   rz'keras.applications.densenet.DenseNet169zkeras.applications.DenseNet169c *[/SQUUUUUUU5$)z*Instantiates the Densenet169 architecture.rmrrs r DenseNet169rsrrz'keras.applications.densenet.DenseNet201zkeras.applications.DenseNet201c *[/SQUUUUUUU5$)z*Instantiates the Densenet201 architecture.rorrs r DenseNet201rrrz,keras.applications.densenet.preprocess_inputc,[R"XSS9$)Ntorch)rOmode)rpreprocess_input)rrOs rrrs  * *  rz.keras.applications.densenet.decode_predictionsc*[R"XS9$)N)top)rdecode_predictions)predsrs rrrs  , ,U <rsI "! 4(2)*:MKK?@ KK?@ KK?@     4 @   #Pf-/O   #,-/O   #,-/O   #,<=> >?=@=*>>EE 55  3 3F ,>>FF4l Y 3 3c 9: Y 3 3c 9: Y 3 3c 9:r